Back to Benchmarks

Claude Opus 4.6 (Adaptive Reasoning, Max Effort) vs Qwen3.5 397B A17B (Reasoning)

Claude Opus 4.6 (Adaptive Reasoning, Max Effort)

Anthropic

vs

Qwen3.5 397B A17B (Reasoning)

Alibaba

Claude Opus 4.6 (Adaptive Reasoning, Max Effort) wins 5 of 11

Index Scores

53.0
+8.0
Intelligence
45.0
48.1
+6.8
Coding
41.3
0.0
Math
0.0

Benchmarks

0.0
MMLU Pro
0.0
89.6
+0.3
GPQA
89.3
36.7
+9.4
HLE
27.3
0.0
LiveCode
0.0
51.9
+9.9
SciCode
42.0
0.0
MATH-500
0.0
0.0
AIME
0.0

Speed

36
TPS
+17
53
Shared
Winning delta

crafted by bart stefanski